WO2020134361A1

WO2020134361A1 - State evaluation method for secondary equipment of substation, system, and equipment

Info

Publication number: WO2020134361A1
Application number: PCT/CN2019/110940
Authority: WO
Inventors: 张越; 张佳楠; 李如意; 单连飞; 吕宏伟; 余建明; 刘艳; 卓峻峰; 张连超
Original assignee: 北京科东电力控制系统有限责任公司
Priority date: 2018-12-27
Filing date: 2019-10-14
Publication date: 2020-07-02
Also published as: CN109697570B; CN109697570A

Abstract

A state evaluation method for secondary equipment of a substation, pertaining to the technical field of secondary substation equipment evaluations. The method comprises: establishing a cloud technology platform (12) for a secondary equipment state, wherein the cloud technology platform (12) comprises a distributed file system module (121) and a state evaluation classifier (122) employing a MapReduce-based extreme learning machine, the distributed file system module (121) reads input data and stores the input data in the form of a secondary equipment state evaluation training set, and the secondary equipment state evaluation training set is applied to training the state evaluation classifier (122) employing the MapReduce-based extreme learning machine, so as to obtain a secondary equipment state evaluation model; and mapping real-time operation data by means of the secondary equipment state evaluation model to directly obtain an output result. The MapReduce-based extreme learning machine is utilized to diagnose and analyze a state of secondary equipment, enables construction of an efficient distributed system having high fault tolerance, high scalability, low costs, and superior extensibility, while also effectively enhancing the capacity to process a large amount of secondary equipment data, and classification accuracy of the classifier.

Description

Method, system and equipment for evaluating secondary equipment state of substation

Technical field

The invention relates to the technical field of secondary equipment evaluation in substations, in particular to a method, system and equipment for evaluating the status of secondary equipment in substations.

Background technique

The safety and reliability of the secondary equipment is a powerful guarantee for the stable operation of the power system. At present, there are few methods for assessing the status of secondary equipment in smart substations. Research is mainly focused on the status monitoring and evaluation of the reliability of relay protection devices. In the operation and maintenance of relay protection systems, the regular maintenance mode of secondary equipment is usually adopted. Such methods may have problems such as "excessive maintenance" or "insufficient maintenance" resulting in unclear equipment health and failure to quickly lock the fault point. With the continuous construction of the smart grid, the problems of data redundancy and low processing efficiency in the mass data processing of the wide-area measurement system, especially the surge of secondary equipment and irregular mass data of secondary equipment inspection results, existing technology Cannot meet the data processing requirements of massive secondary equipment.

Summary of the invention

In view of this, the object of the present invention is to provide a method, system and equipment for evaluating the secondary equipment status of a substation, so as to improve the processing capability of massive secondary equipment data and the classification accuracy of the classifier.

In a first aspect, an embodiment of the present invention provides a method for evaluating the status of secondary equipment in a substation, which is used to evaluate the health status of the secondary equipment in the relay protection system of the intelligent substation;

The data communication module reads the historical operating data of various secondary equipment;

The information extraction module reads the historical operation data in the data communication module and quantizes the corresponding data into processable input data;

Build a secondary device status cloud technology platform. The secondary device status cloud technology platform includes a distributed file system module and a MapReduce-based extreme learning machine state assessment classifier. The distributed file system module reads input data and divides the input data into two. The secondary equipment state assessment training set is stored in form, and the secondary equipment state assessment training set is used to train the MapReduce-based extreme learning machine state assessment classifier to obtain the secondary equipment state assessment model;

Real-time operation data is directly output through the second equipment state assessment model mapping;

The evaluation module evaluates the accuracy of the output result. If the evaluation result is incorrect, the secondary device status evaluation model is revised through the secondary device status cloud technology platform.

Further, the method for the data communication module to read the historical operation data of various secondary devices includes:

The D5000 system collects the real-time operating data of the secondary equipment in real time and stores the historical operating data of the secondary equipment. The data communication module interacts with the server communication software of the D5000 system through the client communication software through the RS485 communication protocol.

Further, the secondary device status cloud technology platform feeds back the correct output result evaluated by the evaluation module to the D5000 system through the data communication module.

Further, the historical operation data of the secondary equipment includes secondary equipment alarm information and maintenance information. The alarm information includes type I alarm and type II alarm; the maintenance information includes the number of refusal to operate, the number of misoperations, family defects, countermeasures implementation, equipment defects 3. The service life.

Further, the input data includes the number of secondary equipment alarms, the specific number of defects and the health status of the secondary equipment. The secondary equipment warning counts include the number of Class I alarms and the number of Class II alarms; , Family defects, countermeasures implementation, equipment defects, service life.

Further, the output result is the health status of the secondary device, and the output result includes normal, attention, abnormal, and severe.

Further, the method for training the state assessment classifier of the extreme learning machine based on MapReduce by using the secondary equipment state assessment training set includes:

Step S1: The input data is stored in the distributed file system module in the form of a secondary device state assessment training set. The input data includes the number of secondary device alarms, the specific number of defects, and the secondary device health status;

Step S2: Read the secondary device status evaluation training set in the distributed file system module, and obtain the training subset of n secondary devices and the training subset of n secondary devices through the underlying mechanism of the distributed computing framework MapReduce The number is consistent with the number of Maps of the secondary device status cloud technology platform;

Step S3: The Map function logic in the MapReduce framework is implemented according to the extreme learning machine algorithm, that is, constructing an extreme learning machine classification model of n secondary device status information, and training the corresponding extreme learning machine classification model through the training subset;

Step S4: The secondary equipment state evaluation results of n Map functions are transmitted to the Reduce function through the shuffle stage in the MapReduce framework, and the Reduce function integrates the secondary equipment state evaluation results to obtain the secondary equipment state evaluation model.

Further, the secondary equipment status assessment model is implemented using a classification model built by an extreme learning machine. Multiple secondary classification models are constructed to achieve multi-classification of secondary equipment status assessment. The model input samples of the classification model are: secondary equipment alarm times and The specific number of defects (type I alarm, type II alarm, number of refusal to operate, number of false operations, family defects, countermeasures implementation, device defects, service life); the output results are: normal, attention, abnormal, serious;

Specific steps are as follows:

Step D1: Set the sample set of secondary device status information

x _j ∈R ^d , y _j ∈{normal, attention, abnormal, serious}, define the extreme learning machine classification model as:

Where: α _i is the weight vector of the i-th hidden layer node at the input node; β _i is the weight vector of the i-th hidden layer node at the output node; b _i is the offset of the i-th hidden layer node; L is the number of hidden layer nodes; G is the activation function of the hidden layer, where the Sigmoid function is taken; N is the number of samples;

Step D2: Solve the classification model of the extreme learning machine, write the matrix as Hβ=T, where the output matrix of the hidden layer of the network is

Step D3: The training process of the samples of the extreme learning machine classification model can be equivalent to finding the least square solution of the equation Hβ=T, the expression is as follows:

In the formula:

The Moore-Penrose generalized inverse of the hidden layer output matrix of the extreme learning machine.

In a second aspect, an embodiment of the present invention also provides a system for evaluating the secondary equipment status of a substation, including:

Data communication module, connect multiple types of secondary equipment and read the historical operation data of the secondary equipment;

The information extraction module extracts the historical operation data of the secondary equipment of the data communication module and quantizes the corresponding data into processable input data;

Secondary device state cloud technology platform, including distributed file system module, and MapReduce-based extreme learning machine state assessment classifier, distributed file system module reads input data and stores the input data in the form of secondary device state assessment training set , The secondary equipment state assessment training set is used to train the MapReduce-based extreme learning machine state assessment classifier to obtain the secondary equipment state assessment model;

Output result module, real-time operating data can be directly output through the secondary equipment state assessment model mapping;

The evaluation module evaluates the accuracy of the output result. If the evaluation result is correct, the output result is the secondary device status evaluation result. If the evaluation result is incorrect, the secondary device status evaluation model is corrected through the secondary device status cloud technology platform.

In a third aspect, an embodiment of the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, when the processor executes the computer program A method for evaluating the status of secondary equipment in substations, including:

The embodiment of the present invention brings the following beneficial effects: The method uses the distributed file system module and the MapReduce-based extreme learning machine state evaluation classifier to train the historical operation data of the secondary device and obtain the secondary device state evaluation model. Real-time real-time operation data is directly output through the secondary equipment state evaluation model mapping, and the output result can be fed back to the dispatching and monitoring room to process the current secondary equipment accordingly, which can improve the processing capacity of massive secondary equipment data And the classification accuracy of the classifier to avoid problems such as unclear secondary equipment health monitoring results and failure to quickly lock the fault point.

Other features and advantages of the present invention will be explained in the subsequent description, and partly become obvious from the description, or be understood by implementing the present invention. The objects and other advantages of the present invention are achieved and obtained by the structures specified in the description, claims and drawings.

In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments are described below in conjunction with the accompanying drawings, which are described in detail below.

BRIEF DESCRIPTION

In order to more clearly explain the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required in the specific embodiments or the description of the prior art. Obviously, the appended The drawings are some embodiments of the present invention. For those of ordinary skill in the art, without paying any creative labor, other drawings can also be obtained based on these drawings.

FIG. 1 is a schematic diagram of a system for evaluating secondary equipment status in a substation according to an embodiment of the present invention;

2 is a schematic diagram of a method for evaluating the state of secondary equipment in a substation according to an embodiment of the present invention;

3 is a schematic diagram of a method for training a state assessment classifier based on a MapReduce-based extreme learning machine using a secondary device state assessment training set according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of a method for implementing a classification model constructed by an extreme learning machine for a secondary device state assessment model provided by an embodiment of the present invention.

icon:

10-Data communication module; 11-Information extraction module; 12-Secondary equipment status cloud technology platform; 121-Distributed file system module; 122-MapReduce-based extreme learning machine state assessment classifier; 13-Output result module; 14 -Evaluation module.

detailed description

To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be described clearly and completely in conjunction with the drawings. Obviously, the described embodiments are part of the embodiments of the present invention, but not all的实施例。 Examples. Based on the embodiments of the present invention, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present invention.

At present, there are few methods for assessing the status of secondary equipment in intelligent substations. Research is mainly focused on the status monitoring and evaluation of the reliability of relay protection devices. In the operation and maintenance of relay protection systems, the regular maintenance mode of secondary equipment is usually adopted. The method may have "excessive maintenance" or "insufficient maintenance", resulting in problems such as unclear equipment health status and failure to quickly lock the fault point. Based on this, the method, system and equipment for secondary equipment status assessment of a substation provided in the embodiments of the present invention may Improve the processing capacity of massive secondary equipment data and the classification accuracy of the classifier.

In order to facilitate the understanding of this embodiment, first, a method for evaluating the state of secondary equipment in a substation disclosed in the embodiments of the present invention will be described in detail.

Example one:

As shown in Figure 1-2, a method for evaluating the status of secondary equipment in a substation is used to evaluate the health status of the secondary equipment in the relay protection system of the intelligent substation; the secondary equipment of power is to monitor the primary equipment in the power system. Auxiliary equipment for measurement, control, protection and regulation; the method includes the following steps:

S101: The data communication module 10 reads the historical operation data of various types of secondary equipment. The historical operation data of the secondary equipment includes secondary equipment alarm information and maintenance information. The alarm information includes Type I alarm and Type II alarm; Number of movements, number of malfunctions, family defects, implementation of countermeasures, equipment defects, service life;

The type I alarm is that when the type I alarm occurs, the relevant plug-ins and modules need to be replaced in time, otherwise the measurement and control and protection functions of the primary device will be lost;

The type II alarm is that when the type II alarm occurs, it can usually be solved by resetting and re-downloading the software operation, and will not pose a direct threat to the operation of the secondary equipment for the time being;

S102: The information extraction module 11 reads the historical operation data in the data communication module 10 and quantizes the corresponding data into processable input data. The input data includes the number of secondary device alarms, the specific number of defects, and the secondary device health status. The number of secondary equipment warnings includes the number of Class I alarms and the number of Class II alarms; the specific number of defects includes the number of refusal to operate, the number of misoperations, family defects, countermeasures, equipment defects, and service life; once the type I alarm occurs, it is necessary to replace the relevant plug-in And module, otherwise it will lose the measurement and control and protection functions of the primary equipment;

S103: Build a secondary device state cloud technology platform 12, the secondary device state cloud technology platform 12 includes a distributed file system module 121 and a MapReduce-based extreme learning machine state assessment classifier 122, and the distributed file system module 121 reads input The data and the input data are stored in the form of a secondary equipment state assessment training set, and the secondary equipment state assessment training set is used to train the MapReduce-based extreme learning machine state assessment classifier 122 to obtain a secondary equipment state assessment model;

Distributed file system (HDFS) to achieve the underlying support of distributed storage;

MapReduce is a program model for processing massive data sets in a cluster, used to solve its distributed storage and computing problems. The MapReduce framework shields the underlying specific implementation details, greatly reducing the difficulty of implementation. Complex parallel computing on large-scale clusters is abstracted into two functions that can be written by users, namely Map and Reduce functions. The specific description is as follows:

1) Input. The input data is first read from the distributed file system, and then it is divided into data pieces. The MapReduce framework allocates a piece of data for each Map function.

2) Map. The MapReduce framework treats input data fragments as a set of (Key, value) key-value pairs. According to the logic of the Map function program written by the user, run and process the (Key, value) key-value pairs assigned by the framework. Finally, a new (Key, value) intermediate key-value pair is generated.

3) Shuffle. At this stage, the intermediate key-value pair is transferred from the Map node to the Reduce node. Depending on the bandwidth and CPU speed, it may take longer than the map and reduce functions to run. At this stage, there is also the work of merging the intermediate values corresponding to the same intermediate key, forming (Key, list of values), and sorting key values.

4) Reduce. Execute the reduce function written by the user. Iterate through all the intermediate values and corresponding intermediate keys or intermediate key chains (list of values), run the data processing logic set by the user, and output new (key, value) key-value pairs.

5) Output. Output the output of Reduce to the specified distributed file system (HDFS) path.

S104: Real-time operation data is directly mapped to the output of the secondary equipment status evaluation model, and the output results include normal, attention, abnormal, and severe;

S105: The evaluation module 14 evaluates the accuracy of the output result. If the evaluation result is not correct, the secondary device state evaluation model is corrected through the secondary device state cloud technology platform 12.

Example two:

A method for evaluating the state of secondary equipment in a substation, which is used to evaluate the health status of the secondary equipment in the relay protection system of an intelligent substation. The method includes the following steps:

S101: The data communication module 10 reads the historical operation data of various secondary devices, and the method of the data communication module 10 reads the historical operation data of various secondary devices includes:

The D5000 system collects real-time operating data of the secondary equipment in real time and stores the historical operating data of the secondary equipment. The data communication module 10 interacts with the server communication software of the D5000 system through the client communication software through the RS485 communication protocol. ;

S102: The information extraction module 11 reads the historical operation data in the data communication module 10 and quantizes the corresponding data into processable input data;

S104: The real-time operation data is directly mapped to the output result through the secondary equipment state assessment model mapping;

S105: The evaluation module 14 evaluates the accuracy of the output result, and if the evaluation result is incorrect, the secondary device status evaluation model is revised through the secondary device status cloud technology platform 12;

S106: The secondary device status cloud technology platform 12 feeds back the output result evaluated by the evaluation module 14 to the D5000 system through the data communication module 10, and the D5000 system can also send the output result instruction to the dispatch monitoring room.

In this embodiment, the secondary equipment status evaluation system is extended to a two-way communication system, which can be used to improve the processing capacity of the massive secondary equipment data and the classification accuracy of the classifier, and can also feed back the results of real-time operating data to the substation in real time. The D5000 system can quickly lock down faults and quickly respond to faults, avoiding the problem of unclear monitoring results of secondary equipment health status.

Example three:

A method for evaluating the state of secondary equipment in a substation, used to evaluate the health status of the secondary equipment in the relay protection system of an intelligent substation;

The data communication module 10 reads the historical operating data of various secondary devices;

The information extraction module 11 reads the historical operation data in the data communication module 10 and quantizes the corresponding data into processable input data;

Build a secondary device state cloud technology platform 12, the secondary device state cloud technology platform 12 includes a distributed file system module 121 and a MapReduce-based extreme learning machine state assessment classifier 122, the distributed file system module 121 reads the input data and The input data is stored in the form of a secondary equipment state assessment training set, and the secondary equipment state assessment training set is used to train the MapReduce-based extreme learning machine state assessment classifier 122 to obtain a secondary equipment state assessment model;

As shown in FIG. 3, the method for applying the secondary device state assessment training set to train the MapReduce-based extreme learning machine state assessment classifier 122 includes:

Step S1: The input data is stored in the distributed file system module 121 in the form of a secondary device state assessment training set. The input data includes the number of secondary device alarms, the specific number of defects, and the secondary device health status;

Step S2: Read the secondary device state assessment training set in the distributed file system module 121, and obtain the training subset of n secondary devices and the training subset of n secondary devices through the underlying mechanism of the distributed computing framework MapReduce The number is the same as the number of Maps in the secondary device status cloud technology platform 12;

Step S4: The secondary equipment state evaluation results of n Map functions are transmitted to the Reduce function through the shuffle stage in the MapReduce framework, and the Reduce function integrates the secondary equipment state evaluation results to obtain the secondary equipment state evaluation model;

The secondary equipment state assessment model is implemented using a classification model built by an extreme learning machine. Multiple secondary classification models are constructed to achieve multi-classification of secondary equipment state assessment. The model input samples of the classification model are: the number of secondary equipment alarms and the specifics of defects Number (Type I alarm, Type II alarm, number of refusal actions, number of misoperations, family defects, countermeasures implementation, device defects, service life); output results: normal, attention, abnormal, serious;

As shown in Figure 4, the specific steps are as follows:

Step D1: Set the sample set of secondary device status information

In the formula:

Moore-Penrose generalized inverse of the hidden layer output matrix of the extreme learning machine;

The evaluation module 14 evaluates the accuracy of the output result, and if the evaluation result is not correct, the secondary device state evaluation model is corrected through the secondary device state cloud technology platform 12.

Embodiment 4: As shown in FIG. 1, a substation secondary equipment condition evaluation system includes:

The data communication module 10 connects multiple types of secondary equipment and reads the historical operation data of the secondary equipment;

The information extraction module 11 extracts the historical operation data of the secondary device of the data communication module 10 and quantizes the corresponding data into processable input data;

The secondary device status cloud technology platform 12 includes a distributed file system module 121 and a MapReduce-based extreme learning machine status evaluation classifier 122. The distributed file system module 121 reads input data and evaluates the input data with the secondary device status The training set is stored in a form, and the secondary equipment state assessment training set is used to train the MapReduce-based extreme learning machine state assessment classifier 122 to obtain a secondary equipment state assessment model;

Output result module 13, real-time running data can be directly output through secondary equipment state assessment model mapping;

The evaluation module 14 evaluates the accuracy of the output result. If the evaluation result is correct, the output result is the secondary device state evaluation result. If the evaluation result is incorrect, the secondary device state evaluation model is corrected through the secondary device state cloud technology platform 12.

Embodiment 5: An electronic device, including a memory, a processor, and a computer program stored on the memory and capable of running on the processor, when the processor executes the computer program, the state of the secondary device of the substation is realized Evaluation methods, including:

The data communication module 10 reads the historical operation data of various secondary equipment;

In the method, system and equipment in the above embodiments, the distributed file system module 121 and the MapReduce-based extreme learning machine state evaluation classifier 122 train the historical operation data of the secondary device and obtain the secondary device state evaluation model. Real-time real-time operation data is directly output through the secondary equipment state evaluation model mapping, and the output result can be fed back to the dispatching and monitoring room to process the current secondary equipment accordingly, which can improve the processing capacity of massive secondary equipment data And the classification accuracy of the classifier to avoid problems such as unclear secondary equipment health monitoring results and failure to quickly lock the fault point.

The block diagram in the drawings shows the possible implementation architecture, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the block diagram may represent a module, program segment, or part of code that contains one or more executable instructions for implementing a specified logical function . It should also be noted that in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two consecutive blocks can actually be executed substantially in parallel, and sometimes they can also be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts, can be implemented with a dedicated hardware-based system that performs specified functions or actions Or, it can be realized by a combination of dedicated hardware and computer instructions.

In addition, in the description of the embodiments of the present invention, unless otherwise clearly specified and defined, the terms "installation", "connection", and "connection" should be understood in a broad sense, for example, it may be a fixed connection or a detachable connection , Or integrally connected; it can be a mechanical connection or an electrical connection; it can be directly connected, or it can be indirectly connected through an intermediary, or it can be the connection between two components. For those of ordinary skill in the art, the specific meaning of the above terms in the present invention can be understood in specific situations.

A computer program product for implementing a method for evaluating the state of a secondary device in a substation when a computer program is provided according to an embodiment of the present invention includes a computer-readable storage medium that stores a non-volatile program code executable by a processor, and the program code includes The instruction of can be used to execute the method described in the foregoing method embodiment. For specific implementation, refer to the method embodiment, which will not be repeated here.

Claims

Substation secondary equipment status assessment method, used to assess the health status of secondary equipment in the relay protection system of intelligent substations, characterized by:

The data communication module reads the historical operating data of various secondary equipment;

The information extraction module reads the historical operation data in the data communication module and quantizes the corresponding data into processable input data;

Build a secondary device status cloud technology platform. The secondary device status cloud technology platform includes a distributed file system module and a MapReduce-based extreme learning machine state assessment classifier. The distributed file system module reads input data and divides the input data into two. The secondary equipment state assessment training set is stored in form, and the secondary equipment state assessment training set is used to train the MapReduce-based extreme learning machine state assessment classifier to obtain the secondary equipment state assessment model;

Real-time operation data is directly output through the second equipment state assessment model mapping;

The evaluation module evaluates the accuracy of the output result. If the evaluation result is incorrect, the secondary device status evaluation model is revised through the secondary device status cloud technology platform.
The method for evaluating the status of secondary equipment in a substation according to claim 1, wherein the method for the data communication module to read the historical operation data of various types of secondary equipment includes:

The D5000 system collects the real-time operating data of the secondary equipment in real time and stores the historical operating data of the secondary equipment. The data communication module interacts with the server communication software of the D5000 system through the client communication software through the RS485 communication protocol.
According to claim 2, characterized in that the secondary device status cloud technology platform feeds back the correct output result evaluated by the evaluation module to the D5000 system through the data communication module.
The method for evaluating the status of secondary equipment in a substation according to claim 1, wherein the historical operation data of the secondary equipment includes secondary equipment alarm information and maintenance information, and the alarm information includes type I alarm and type II alarm; the maintenance information includes Number of refusal to operate, number of misoperations, family defects, implementation of countermeasures, equipment defects, service life.
The method according to claim 1, wherein the input data includes the number of secondary device alarms, the specific number of defects and the health status of the secondary device, and the number of secondary device warnings includes the number of Class I alarms and the number of Class II alarms; the specific defects The number includes the number of rejections, the number of misoperations, family defects, implementation of countermeasures, equipment defects, and years of use.
The method according to claim 1, wherein the output result is a secondary device health state, and the output result includes normal, attention, abnormal, and severe.
The method according to claim 1, wherein the method for applying a secondary device state assessment training set to train a MapReduce-based extreme learning machine state assessment classifier includes:

The input data is stored in the distributed file system module in the form of a secondary device state assessment training set. The input data includes the number of secondary device alarms, the specific number of defects, and the secondary device health status;

Read the secondary device state assessment training set in the distributed file system module, and obtain the training subset of n secondary devices through the underlying mechanism of the distributed computing framework MapReduce. The number and training sub-sets of n secondary devices The number of Maps of the cloud technology platform of the secondary device status is consistent;

The Map function logic in the MapReduce framework is implemented based on the extreme learning machine algorithm, that is, constructing an extreme learning machine classification model of n secondary device status information, and training the corresponding extreme learning machine classification model through the training subset;

The secondary equipment state evaluation results of n Map functions are transmitted to the Reduce function through the shuffle stage in the MapReduce framework, and the Reduce function integrates the secondary equipment state evaluation results to obtain the secondary equipment state evaluation model.
According to claim 7, characterized in that, the secondary equipment state assessment model is implemented using a classification model constructed by an extreme learning machine, and multiple classifications of secondary equipment state assessment are realized by constructing multiple binary classification models, and the model input of the classification model The sample is: the number of secondary equipment alarms and the specific number of defects (type I alarm, type II alarm, number of refusal to operate, number of false operations, family defects, implementation of countermeasures, device defects, service life); the output results are: normal, attention , Abnormal, serious;

Specific steps are as follows:

Set a sample set of secondary device status information
x j ∈R d , y j ∈{normal, attention, abnormal, serious}, define the extreme learning machine classification model as:

Where: α i is the weight vector of the i-th hidden layer node at the input node; β i is the weight vector of the i-th hidden layer node at the output node; b i is the offset of the i-th hidden layer node; L is the number of hidden layer nodes; G is the activation function of the hidden layer, where the Sigmoid function is taken; N is the number of samples;

Solve the classification model of the extreme learning machine and write it as a matrix with Hβ = T, where the output matrix of the hidden layer of the network is

The training process of the samples of the extreme learning machine classification model can be equivalent to finding the least square solution of the equation Hβ=T, and the expression is as follows:

In the formula:
The Moore-Penrose generalized inverse of the hidden layer output matrix of the extreme learning machine.
A state assessment system for secondary equipment in a substation, including:

Data communication module, connect multiple types of secondary equipment and read the historical operation data of the secondary equipment;

The information extraction module extracts the historical operation data of the secondary equipment of the data communication module and quantizes the corresponding data into processable input data;

Secondary device state cloud technology platform, including distributed file system module, and MapReduce-based extreme learning machine state assessment classifier, distributed file system module reads input data and stores the input data in the form of secondary device state assessment training set , The secondary equipment state assessment training set is used to train the MapReduce-based extreme learning machine state assessment classifier to obtain the secondary equipment state assessment model;

Output result module, real-time operating data can be directly output through the secondary equipment state assessment model mapping;

The evaluation module evaluates the accuracy of the output result. If the evaluation result is correct, the output result is the secondary device status evaluation result. If the evaluation result is incorrect, the secondary device status evaluation model is corrected through the secondary device status cloud technology platform.
An electronic device, including a memory, a processor, and a computer program stored on the memory and runable on the processor, characterized in that, when the processor executes the computer program, claims 1 to 8 are implemented The method of any one of the steps.